Stress fractures are an important clinical manifestation of fatigue failure in bone. In addition, there is growing speculation that fatigue damage is an important component of bone fragility in the elderly. The goal of this study is to develop a quantitative understanding of the repair of fatigue damage in cortical bone by osteonal remodeling. Previous experimental work has shown that a quantum of fatigue damage, manifest by the appearance of histologically observable microcracks, may be introduced by bending a canine radius and ulna repetitively in vivo, and that this initiates additional remodeling. Many investigators have hypothesized that such a remodeling response is induced by, and intended to repair, the fatigue damage. However, the degree of damage removal by the incipient additional remodeling, the time required, and its relationship to the remodeling variables, all remain unstudied. This is the work that the applicants propose to undertake, using a computer model to develop the theoretical concepts, and the same in vivo canine experimental model to test and refine the theory. Experiment 1 will use 20 dogs to determine the relationship between the magnitude of the applied strain and crack damage. Experiment 2 will use 24 dogs to develop a dose-response relationship between crack damage and increased remodeling. Four strain magnitudes and 3 response times will be studied using a complete block ANOVA design. Experiment 3 will use 16 dogs to test whether or not the induced remodeling actually removes the induced damage, and test a theoretical model for the factors governing the rate of removal. an ANOVA design will be used here as well, with 2 damage levels and 4 repair times. Another important factor in the problem of fatigue damage repair is the possible interaction between damage-induced remodeling and subsequent damage accumulation. Increased remodeling introduces additional porosity, which may exacerbate the damage produced by continued loading. Experiment 4 will use 10 dogs to test for the existence of such an interaction, and the ability of the theoretical model to predict its effects. In this case, one limb will be loaded to induce remodeling, then loaded again after increased porosity has had time to develop. At this time the contralateral control limb will receive the combined loadings, and the bones will be harvested and analyzed. In all these experiments, en bloc staining will be used to identify fatigue cracks, and conventional histomorphometric analysis with fluorochrome labels will be used to quantify bone remodeling and its effects on porosity.